That is a reasonable suggestion to me. There is no guarantee that covariates are balanced
w.r.t. the original treatment when the balance was achieved with the dichotomized
treatment, but the balance is probably improved.
Kosuke
On Mar 9, 2011, at 2:46 PM, Casey Klofstad wrote:
Colleagues-
In a recent review I received, it was suggested that I match using the
dichotomous measure of the treatment variable (as is, of course,
necessary), but then run the weighted regression with the full scale
of the treatment variable, rather than then dichotomous version. Is
this kosher, or one must one still use the dichotomous version of the
treatment variable in the final analysis?
Thanks much,
-c
--
Casey A. Klofstad
University of Miami, Political Science
http://www.as.miami.edu/personal/cklofstad
************
Please check out my book at:
http://www.temple.edu/tempress/titles/2099_reg.html
--
MatchIt mailing list served by Harvard-MIT Data Center
List Address: matchit(a)lists.gking.harvard.edu
Subscribe/Unsubscribe:
http://lists.gking.harvard.edu/?info=matchit
MatchIt Software and Documentation:
http://gking.harvard.edu/matchit/
Browse/Search List Archive:
http://lists.hmdc.harvard.edu/lists/matchit/
--
MatchIt mailing list served by Harvard-MIT Data Center
List Address: matchit(a)lists.gking.harvard.edu
Subscribe/Unsubscribe:
http://lists.gking.harvard.edu/?info=matchit
MatchIt Software and Documentation:
http://gking.harvard.edu/matchit/
Browse/Search List Archive:
http://lists.hmdc.harvard.edu/lists/matchit/